[R-sig-ME] [FORGED] cloglog models (Rolf's question from a few days ago)

Paul Buerkner paul.buerkner at gmail.com
Sat Jul 22 22:03:12 CEST 2017


brms shows some strange behavior as well. I will look into it and see what
can be improved.

2017-07-22 21:58 GMT+02:00 Douglas Bates <bates at stat.wisc.edu>:

> Just a note about the Julia version.  Ben and I have been discussing it in
> https://github.com/dmbates/MixedModels.jl/issues/92.  This example turned
> up a couple of issues with the code.  Having fixed those the Julia version
> is now the one of the fastest and produces a log-likelihood value in excess
> of the others.  Ben says he will modify his writeup accordingly.
>
> On Fri, Jul 21, 2017 at 12:09 AM Rolf Turner <r.turner at auckland.ac.nz>
> wrote:
>
> >
> > On 21/07/17 14:53, Ben Bolker wrote:
> >
> > >
> > >    I got a little carried away.
> > >
> > >   cloglog examples here with various glmer settings, glmmTMB,
> > > MixedModels.jl (thanks to Doug Bates for troubleshooting) ...
> > >
> > >   http://bbolker.github.io/mixedmodels-misc/notes/cloglogsim.html
> > >
> > >
> > https://github.com/bbolker/mixedmodels-misc/blob/master/
> notes/cloglogsim.rmd
> >
> > Dear Ben,
> >
> > Holy Ramoles!!!  Did you *ever* get carried away!  There's a lot there
> > for me to digest; probably deeper than I am *able* to digest, but I'll
> try.
> >
> > Thanks hugely for the time and effort that you've put into this.  I hope
> > that it was rewarding from your pov.
> >
> > I will very likely be coming back to you/this list with more questions,
> > once I've thought things over for a while.
> >
> > Thanks again.
> >
> > cheers,
> >
> > Rolf
> >
> > --
> > Technical Editor ANZJS
> > Department of Statistics
> > University of Auckland
> > Phone: +64-9-373-7599 ext. 88276 <+64%209-373%207599>
> >
> > _______________________________________________
> > R-sig-mixed-models at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
> >
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>

	[[alternative HTML version deleted]]



More information about the R-sig-mixed-models mailing list